Add deffered compilation for render and compute pipelines

* Add `pipeline_deffered_rd.h` for the motion blur PR

Author: AR-DEV-1

servers/rendering/renderer_rd/pipeline_deferred_rd.h

@@ -0,0 +1,135 @@
+/**************************************************************************/
+/*  pipeline_deferred_rd.h                                                */
+/**************************************************************************/
+/*                         This file is part of:                          */
+/*                             REDOT ENGINE                               */
+/*                        https://redotengine.org                         */
+/**************************************************************************/
+/* Copyright (c) 2024-present Redot Engine contributors                   */
+/*                                          (see REDOT_AUTHORS.md)        */
+/* Copyright (c) 2014-present Godot Engine contributors (see AUTHORS.md). */
+/* Copyright (c) 2007-2014 Juan Linietsky, Ariel Manzur.                  */
+/*                                                                        */
+/* Permission is hereby granted, free of charge, to any person obtaining  */
+/* a copy of this software and associated documentation files (the        */
+/* "Software"), to deal in the Software without restriction, including    */
+/* without limitation the rights to use, copy, modify, merge, publish,    */
+/* distribute, sublicense, and/or sell copies of the Software, and to     */
+/* permit persons to whom the Software is furnished to do so, subject to  */
+/* the following conditions:                                              */
+/*                                                                        */
+/* The above copyright notice and this permission notice shall be         */
+/* included in all copies or substantial portions of the Software.        */
+/*                                                                        */
+/* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,        */
+/* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF     */
+/* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. */
+/* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY   */
+/* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,   */
+/* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE      */
+/* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.                 */
+/**************************************************************************/
+
+#pragma once
+
+#include "servers/rendering/rendering_device.h"
+
+// Helper class for automatically deferring compilation of a pipeline to a background task.
+// When attempting to retrieve the pipeline with the getter, the caller will automatically
+// wait for it to be ready.
+
+class PipelineDeferredRD {
+protected:
+	struct CreationParameters {
+		RID shader;
+		RD::FramebufferFormatID framebuffer_format;
+		RD::VertexFormatID vertex_format;
+		RD::RenderPrimitive render_primitive;
+		RD::PipelineRasterizationState rasterization_state;
+		RD::PipelineMultisampleState multisample_state;
+		RD::PipelineDepthStencilState depth_stencil_state;
+		RD::PipelineColorBlendState blend_state;
+		BitField<RD::PipelineDynamicStateFlags> dynamic_state_flags;
+		uint32_t for_render_pass;
+		Vector<RD::PipelineSpecializationConstant> specialization_constants;
+		bool is_compute;
+	};
+
+	RID pipeline;
+	WorkerThreadPool::TaskID task = WorkerThreadPool::INVALID_TASK_ID;
+
+	void _create(const CreationParameters &c) {
+		if (c.is_compute) {
+			pipeline = RD::get_singleton()->compute_pipeline_create(c.shader, c.specialization_constants);
+		} else {
+			pipeline = RD::get_singleton()->render_pipeline_create(c.shader, c.framebuffer_format, c.vertex_format, c.render_primitive, c.rasterization_state, c.multisample_state, c.depth_stencil_state, c.blend_state, c.dynamic_state_flags, c.for_render_pass, c.specialization_constants);
+		}
+	}
+
+	void _start(const CreationParameters &c) {
+		free();
+		task = WorkerThreadPool::get_singleton()->add_template_task(this, &PipelineDeferredRD::_create, c, true, "PipelineCompilation");
+	}
+
+	void _wait() {
+		if (task != WorkerThreadPool::INVALID_TASK_ID) {
+			const Error err = WorkerThreadPool::get_singleton()->wait_for_task_completion(task);
+			ERR_FAIL_COND_MSG(err != OK, "Failed to wait for pipeline compilation task completion.");
+			task = WorkerThreadPool::INVALID_TASK_ID;
+		}
+	}
+
+public:
+	PipelineDeferredRD() {
+		// Default constructor.
+	}
+
+	~PipelineDeferredRD() {
+		_wait(); // Block until worker thread completes
+#ifdef DEV_ENABLED
+		ERR_FAIL_COND_MSG(pipeline.is_valid(), "'free()' must be called manually before deconstruction and before the corresponding shader is freed.");
+#endif
+	}
+
+	void create_render_pipeline(RID p_shader, RD::FramebufferFormatID p_framebuffer_format, RD::VertexFormatID p_vertex_format, RD::RenderPrimitive p_render_primitive, const RD::PipelineRasterizationState &p_rasterization_state, const RD::PipelineMultisampleState &p_multisample_state, const RD::PipelineDepthStencilState &p_depth_stencil_state, const RD::PipelineColorBlendState &p_blend_state, BitField<RD::PipelineDynamicStateFlags> p_dynamic_state_flags = 0, uint32_t p_for_render_pass = 0, const Vector<RD::PipelineSpecializationConstant> &p_specialization_constants = Vector<RD::PipelineSpecializationConstant>()) {
+		CreationParameters c;
+		c.shader = p_shader;
+		c.framebuffer_format = p_framebuffer_format;
+		c.vertex_format = p_vertex_format;
+		c.render_primitive = p_render_primitive;
+		c.rasterization_state = p_rasterization_state;
+		c.multisample_state = p_multisample_state;
+		c.depth_stencil_state = p_depth_stencil_state;
+		c.blend_state = p_blend_state;
+		c.dynamic_state_flags = p_dynamic_state_flags;
+		c.for_render_pass = p_for_render_pass;
+		c.specialization_constants = p_specialization_constants;
+		c.is_compute = false;
+		_start(c);
+	}
+
+	void create_compute_pipeline(RID p_shader, const Vector<RD::PipelineSpecializationConstant> &p_specialization_constants = Vector<RD::PipelineSpecializationConstant>()) {
+		CreationParameters c = {};
+		c.shader = p_shader;
+		c.specialization_constants = p_specialization_constants;
+		c.is_compute = true;
+		_start(c);
+	}
+
+	RID get_rid() {
+		_wait();
+		return pipeline;
+	}
+
+	void free() {
+		_wait();
+
+		if (pipeline.is_valid()) {
+#ifdef DEV_ENABLED
+			ERR_FAIL_COND_MSG(!(RD::get_singleton()->render_pipeline_is_valid(pipeline) || RD::get_singleton()->compute_pipeline_is_valid(pipeline)), "`free()` must be called manually before the dependent shader is freed.");
+#endif
+			RD::get_singleton()->free_rid(pipeline);
+			pipeline = RID();
+		}
+	}
+};